Spinal cord injury (SCI) results in muscle paralysis causing respiratory muscle dysfunction. In SCI, there is risk for premature death, and diseases of the respiratory system are the most common cause. Previous cross-sectional studies in SCI have design and methodological flaws and, since there has never been a longitudinal assessment, the natural history of respiratory dysfunction following SCI is largely unknown. Specifically unknown, for example, are whether after injury there is an accelerated loss of function greater than expected for aging and whether, other than the injury itself, factors such as neurological level and completeness, time since injury, smoking, and wheeze are independent predictors of deterioration. To study this, from 1994-1998, we recruited a unique cross-sectional cohort of 360 subjects with SCI, the largest such study conducted to date. Additional recruitment and longitudinal follow-up of this cohort began in late 1999 and is currently funded through 6/2002. Although it is common clinical practice to attribute respiratory dysfunction in SCI solely to neurological level and completeness, our cross-sectional analyses suggest that chest illness and pulmonary dysfunction are also related independently to factors some of which are more important than the injury per se. These factors include wheeze (as reported on a standardized respiratory questionnaire), cigarette smoking, previously diagnosed lung and heart disease, and impairment of respiratory muscle strength. Preliminary longitudinal analyses suggest that the effect of continued smoking on pulmonary function is greater in SCI than in the able-bodied. In this application, we propose an additional 5 years of study, starting in 2002, of 390 subjects with SCI. The study will assess the importance of these and other factors in the longitudinal change in pulmonary function and occurrence of chest illness, and will identify potentially modifiable factors that, if modified, may reduce respiratory disease morbidity and mortality in chronic SCI.